Early Language Acquisition and Bilinguilism

The adult brain is a highly specialized machine that responds specifically to different types of stimulation. However, we are not born with such a well-fitted machine and its specialization is the result of the interplay of both internal specifications and environmental exposure. This selective specialization is clearly observed in language acquisition and use. Indeed, from zero to twelve months infants go through a period where they are just able to recognize some broad (rhythmic) characteristics of the maternal language, to having isolated some words from the continuous speech signal and given a meaning to these acoustic patterns. This acquisition of knowledge of the maternal language is accompanied by the loss of certain capacities present at birth. As adults we are extremely efficient at processing our maternal language, and quite poor at processing foreign languages. As the well-known example of difficulties by native Japanese speakers to perceive the English /r/-/l/ contrast shows, even simple phoneme discrimination tasks can be very difficult to perform in a foreign language. However, all newborn infants, Japanese too, are able to correctly perceive that contrast.

In the past twenty years, researchers working in very early language acquisition have done important discoveries in the way this specialization process takes place. However, almost nothing is known of how bilingual infants learn their two languages [the term "bilingual" will refer to individuals who are exposed to two languages from the very first day of their lives, and any other type of "bilingual" exposure will be considered as "L2 acquisition"]. Bilingual infants receive a language input that differs from the monolingual one in some important aspects. One aspect is that of the amount of exposure: although the global exposure maybe the same, the input for each language is roughly half of that of the monolingual infant. The other aspect is that the input is "noisier", since in many occasions the infant gets one language mixed with the other one. Therefore, it is reasonable to expect some differences in the way bilinguals represent their two languages (as compared with monolinguals).

What are the long-term and short-term consequences of this different exposure? Parents rising children in bilingual environments usually ask if it is good or it is bad to do so. Cognitive developmental neuroscience can provide no answer to this question. We know very little both about how bilingual infants acquire their two languages and how adult bilinguals process and represent their two languages in the brain (with very few exceptions, almost all psycholinguistic and brain imaging studies -as well as bilingual aphasia cases- refer to L2 learners). If the way the brain establishes many neuronal connections is done on the basis of the incoming stimulation (by selecting those connections more frequently used), it is highly possible that bilinguals have a different pattern of connections in the language areas of the brain, when compared with monolinguals. Daily experience shows no particular difficulty in adult bilinguals when using any of their two languages, but it could be the case that they use different procedures (or make use of compensatory strategies) that make the differences unnoticeable, unless specific laboratory tools are employed.

For many years, bilingualism has been considered an atypical situation, restricted to the case of immigrants or to a situation confined to a few countries in the world. The current multilingual situation, both in the US and in Europe has put this issue in the front page. Studying bilinguals is learning how an increasing number of human beings acquire language.

The goal of the present project is to gather deeper knowledge of how bilingual infants learn their two languages and, by doing so, we also hope to be able to better understand the mechanisms of brain and language acquisition.

We plan to carry out a research program that will explore how bilingual infants raised in two different bilingual environments acquire their two languages. One potentially important remark is that not all pairs of languages are equally similar. As adults, we find some languages easier to be learnt than others. For instance, for a native Dutch speaker, learning English is relatively easier than learning Japanese. Although the learning situation is completely different for infants, it could be the case that also for them learning two similar languages implies different mechanisms than learning two different languages. Indeed, past research has shown that newborns can notice differences between some pairs of languages, but not between others. In fact, even five-month-olds cannot discriminate some languages (Dutch and German) if they are not their maternal one. The first thing a bilingual infant has to do to learn the two languages of her environment is to notice that there are two languages and not just one. Obviously, being able to notice the existence of two languages at birth or after a few months makes a quite different stating point. The first months of life is a period characterized by intense brain reorganization. As we said, before the end of the first year of life infants have already learnt a few words of their maternal language. In this context, a difference of a few weeks or months in making the distinction between the two languages of the environment may have crucial consequences in later brain development.

Up to this point we have been considering language as if it was a unitary entity. Language is a complex system and learning it implies the acquisition of its sound patterns (the phonology of the language), the morphology and its syntax and semantics. These different parts of the language system are learnt at different moments and they impose restrictions on each other. In this project we plan to study how bilingual infants acquire and represent in the brain these different subtypes of linguistic knowledge We also intend to analyze the long-term consequences of being raised in a bilingual environment (for each of these linguistic domains) by studying adult bilinguals.

One problem that many studies on bilingualism encounter is that of a proper comparison across populations and materials. Indeed, in many studies different languages are compared or individuals from different cultural backgrounds are contrasted. Quite often, if differences between two groups are found, they can be due to some language specific properties or to differences in the participants' cultural background. Furthermore, to successfully carry out an extensive project on bilingual infant language acquisition requires having access to a large and cooperative bilingual population. Thus, a lot of constraints exist in terms of properties of the languages being tested and of the populations under study for a project like the one proposed here.

Spain presents a perfect natural setting to address these issues. There is a single common official language (Spanish) and in different parts of the country, other languages are also spoken (and they are co-official with Spanish). One of the co-official languages is Catalan (spoken in Catalonia by approximately 6 million people). Catalan, as well as Spanish, Italian, Portuguese and French, is a Romance language. In the same way as Italian and Spanish are quite similar, Catalan and Spanish share many characteristics. Another co-official language is Basque (spoken in the Basque Country by approximately x million people). Basque is a non-Indoeuropean language, thus it is very different from all Germanic and Romance languages. Spanish, Catalan and Basque share the same social status and they are present in the media and at all levels of the educational system. Summarizing, monolingual Spanish, bilingual Catalan-Spanish and bilingual Basque-Spanish share a common language and a common cultural background [Catalan and Basque are also spoken in some areas of France]. These circumstances eliminate most of the methodological problems that other studies could have. By studying these two cases of bilingualism we hope to be able to cover two extreme comparisons in the language similarity-dissimilarity continuum. Therefore, the results of the present project should be generalizable to other bilingual situations and language comparisons.

Finally, the results of the present research may serve as a basis for future advice in educational matters. Not only more precise answers concerning the implications of an exposure in early language acquisition could be given, but also some of the consequences of a bilingual situation for children with learning difficulties might be identified.